A Roadmap for Assessing the Quality of Experience of 3D Audio Binaural Rendering

Today there are 2 major evolutions in spatial audio. First, an enhanced 3D audio experience, where virtual sound sources can be accurately synthesized in any direction, is possible with technologies such as binaural, Wave Field Synthesis, Higher Order Ambisonics or Vector Base Amplitude Panning. Second, 3D audio is on the way to being democratized through binaural adaptation for headphone listening. These evolutions call for revisiting the methods and tools used to assess the perception of spatial sound reproduction. The first objective of this paper is to delineate the problem, by exploring the potential dimensions and the related attributes underlying the perception of spatial sound, mainly within the context of binaural reproduction. Secondly, assessment methods, including both standard and less conventional ones, are listed, and their relevance for the measure of the attributes previously identified is discussed.